Quality and Trust Indicators of Digital Road Infrastructure Data Are

Essential to Improve Its Usability: An Intelligent Speed Assist (ISA)

Study

Jacco van de Sluis

1 a

, Daniel Altgassen

2 b

and Peter-Paul Schackmann

Networks department, TNO - ICT, Strategy & Policy, The Hague, The Netherlands

Integrated Vehicle Safety Department, TNO - Mobility & Build Environment, Helmond, The Netherlands

Keywords:

Intelligent Speed Assist (ISA), Misbehaviour Detection and Reporting (MBD&R), Digital Road Infrastructure

(DRI), Vehicle-to-Everything (V2X) Communication, Intelligent Transport Systems (ITS), Data Trust, Data

Quality, Cooperative Connected and Automated Mobility (CCAM), Advanced Driver Assistance System

(ADAS), Automated Driving System (ADS).

Abstract:

The promise of a Digital Road Infrastructure (DRI) is to improve both road and vehicle safety. ADAS/ADS

with DRI support, can help vehicles in overcoming certain sensor limitations, handle more complex opera-

tional situations and offer additional situational awareness. An effective DRI must be trusted and must offer

the required data quality, both are currently lacking. Intelligent Speed Assist (ISA) is an interesting show case

for the added value of DRI. In our approach camera-based trafﬁc sign detections and map-based speed limit

information, both occasionally wrong, are augmented with actual speed limit and road layout information

coming from DRI. A Misbehaviour Detection and Reporting (MBD&R) concept tailored to the ISA sources

is deployed in the vehicle to detect and report ISA related misbehaviour. Trust and quality indicators are

calculated for data coming from camera, map and DRI, which are used to verify and compare theses sources

and make improved ISA speed limit decisions. The vehicle implementation is tested under real-life trafﬁc

conditions. Our work is a ﬁrst step in realizing a trusted DRI. The long-term goal is collaboration among all

stakeholders to implement mechanisms that improve trust and the quality of shared data sources for use in

trafﬁc safety applications.

1 INTRODUCTION

Connected, Cooperative, and Automated Mobility

(CCAM) applications can improve the performance

of existing Advanced Driver Assistance Systems

(ADAS) / Automated Driving Systems (ADS). These

new data-driven services enabled by connectivity can

increase vehicle intelligence, achieve higher levels

of automation and improve safety, see reports from

(Farah et al., 2018) and (OECD, 2023). But to work in

practice, a trusted and shared Digital Road Infrastruc-

ture (DRI) is essential. With numerous components

required to make this possible, it is crucial that vehicle

OEMs, industry partners, service providers, and road

operators collaborate to enhance both road and vehi-

cle safety. An effective DRI offers trusted sources and

offers the required data quality, both of which are cur-

https://orcid.org/0000-0002-7162-6014

https://orcid.org/0009-0003-2813-0499

rently lacking. As our mobility system prepares for

vehicles with higher levels of automation, the need

for a reliable DRI increases. Information generated

by road operators, road users or other data providers -

and shared with speciﬁc road users - can greatly con-

tribute to that goal. From a vehicle perspective, us-

ing external data, on top of the vehicles’ own sensor

data can signiﬁcantly extend the vehicle operational

horizon and thereby improve vehicle safety (Zhang

et al., 2022). Despite DRI potential beneﬁts, the us-

ability of external digital data for safety applications

is low, due to concerns about reliability (quality and

trust) and possible misbehaviours, see (van der Heij-

den et al., 2019) and (Kamel et al., 2020). Also road

operators and data providers are unsure of the speciﬁc

data requirements needed and how to provide these.

To move forward, cross-chain collaboration between

all stakeholders is required, as well as proof that qual-

ity and trust issues can be resolved. This paper uses

the Intelligent Speed Assist (ISA) as study case to fur-

van de Sluis, J., Altgassen, D. and Schackmann, P.-P.

Quality and Trust Indicators of Digital Road Infrastructure Data Are Essential to Improve Its Usability: An Intelligent Speed Assist (ISA) Study.

DOI: 10.5220/0013470100003941

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 11th International Conference on Vehicle Technology and Intelligent Transport Systems (VEHITS 2025), pages 631-637

ISBN: 978-989-758-745-0; ISSN: 2184-495X

631

ther describe the challenges and to introduce possible

solutions.

2 IMPROVING SHARED DATA

USABILITY FOR ISA

Real-time trafﬁc information data is often only used to

inform drivers. In these systems, the driver is respon-

sible for safe driving behavior and occasional misin-

formation is of lesser criticality, as it is handled by

human interpretations. The urgency to decrease mis-

information has recently grown, as the same data is

now also being used for trafﬁc and road safety ap-

plications, such as usage in ADAS. This develop-

ment increases the importance of the detection of mis-

behaviour and having a real-time measure for data

quality and data trust. To properly scale ADAS and

achieve higher levels of automation, relying solely

on ego vehicle sensor data is challenging. Especially

when operational design domains extend and become

more complex, e.g. by covering (sub)urban environ-

ments, diverse road and weather conditions. Cur-

rent ISA vehicle implementations often use a camera

to recognize speed limit signs, combined with map

data to determine the applicable speed limit. How-

ever, camera-based detection can have (limitation)

signs that are obstructed, misinterpreted, or affected

by weather conditions. Additionally, map data can be

inaccurate, insufﬁciently detailed (e.g. missing infor-

mation related to: time-of-day, vehicle-type, weather

conditions), or outdated. These sensor and map ﬂaws

can lead to unsafe situations, such as speeding or un-

expected braking, especially when used by automated

vehicle systems. The EU 2019/2144 ISA regulation

(EC, 2023) makes ISA mandatory in new vehicles. It

does not constrain manufacturers by specifying which

sources to use for speed limit data, but it outlines per-

formance requirements and implementation possibil-

ities. The additional use of digital road infrastruc-

ture (DRI) data can provide the technical solution to

achieve the required ISA performance.

2.1 Misbehaviour Detection

But with increased connectivity, cyber-security be-

comes more apparent. Cyber-attacks and misbe-

haviour are not limited to the connected vehicles, they

need to be considered within the full CCAM ecosys-

tem. The ISA use case is a complex example of a

cyber-physical system-of-systems, attacks and misbe-

haviours can be considered at vehicle level, within

the DRI and data sharing environments. For that

reason, it is important to have Misbehaviour Detec-

Misbehaviour Detection System (@ Vehicle; RSU), Central

Event CategorizationDetectors

Inputs

Alg. 1

Alg. 2

Alg. N

...

Misbehaviour Reporting System (@ Vehicle; RSU, Central)

Decision Generation Transmission

Backend Misbehaviour Management (@ Central Authority)

Revocation Misbehaviour AnalysisDissemination

Security Protocols

Protect

Detect

Respond

Perception

Communication

Report

HD Map +

Localization

IVIM

CAM

CPM

DENM

Trust Score

Report

Intelligent Speed Assistance

Quality

Score

Speed Set Point

Figure 1: MBD&R high-level architecture.

tion (MBD) functionalities available at all these levels

(5GAA, 2022). For the mitigation strategy, it is neces-

sary to have Misbehaviour Reporting Systems (MRS)

to exchange misbehaviour information between rele-

vant entities in the CCAM ecosystem (ETSI, 2023), as

part of the response and recovery steps. A basic Mis-

behaviour Detection and Reporting (MBD&R) setup

is illustrated in Fig. 1. This generic architecture is

taken from our previous work (Oliveira, 2024), which

is also based on the guidelines described in the Mis-

behaviour Detection white paper from the 5GAA Au-

tomotive Association (5GAA, 2022). In-vehicle mis-

behaviour detectors are deployed to check incoming

sensor data and communicated data from DRI, before

its usage in ISA. MRS functions are in place to report

and share detected misbehaviours across the CCAM

ecosystem.

2.2 Data Trust and Data Quality

Standards and trust mechanisms are essential build-

ing blocks to fully realize smart mobility beneﬁts.

With the European Union C-ITS Security Credential

Management System policies (EC, 2018), the EU has

established a European Trust Domain that provides

electronic signatures to ensure origin and integrity of

data (ETSI, 2021). This electronic signature indi-

cates that the shared data has not been changed or

tampered with and that it is from a trusted source.

With this, vehicles can recognize trusted data sources

which is crucial in the assessment of the reliabil-

ity of shared information. This is an important ﬁrst

step, but it is not enough to create a trusted DRI. En-

surance of the right data quality level is crucial as

VEHITS 2025 - 11th International Conference on Vehicle Technology and Intelligent Transport Systems

632

well. Even when data providers are trusted, the data

itself can still be ﬂawed, with errors such as incor-

rect signals or outdated information. Our proposed

solution for this is a Misbehaviour Detection (MBD)

concept to help identify such issues by comparing

multiple data sources and using historical patterns.

Data quality from different sources is operationally

assessed through data sanity checks and comparisons

across these sources. Having both quality and trust

metrics enables the ADAS/ADS system to make in-

formed decisions based on the shared data. The Mis-

behaviour Detection mechanisms can also be used to

report suspected faulty data back to the source, allow-

ing that party to make improvements and thus contin-

uously enhance data quality. The requirements and

complexity of trust and MBD algorithms depend on

the required safety levels. For vehicle systems, in-

creasing levels of automation, means reliability and

data quality become more critical.

3 ISA USE CASE WITH DRI

SUPPORT

In the Digital Infrastructure for Future-proof Mobil-

ity project (DITM, 2022), digital infrastructure for

automated transport solutions are being implemented

into validation labs. ISA supported by DRI is an

important use case of the validation labs. In DITM

MBD&R functions are being developed and imple-

mented at vehicle, roadside and central level. The ve-

hicle deployments use TNO carlabs to evaluate the

ISA MBD&R system under real-life conditions at

public roads. These carlabs are our CCAM research

facilities consisting of regular production vehicles

equipped with relevant retroﬁtted hardware and soft-

ware to enable experimental automated driving appli-

cations. The systems and software are under our own

control and include sensor-sets (camera, radar, lidar,

etc.), wireless communications (ITS-G5, C-V2X, 5G

cellular, UWB, etc.), our ADAS/ADS driving stack

(perception, localization, sensor-fusion, drive plan-

ning, vehicle control, etc.), human-machine interface

(HMI, e.g. for warnings, handing over of control).

In the remaining part of the paper the focus is on

these in-vehicle MBD functions. A generic approach

is used that scales with available sensor-suite, sup-

ports different data sources as multiple types of ap-

plications (trafﬁc safety, ADAS/ADS or other CCAM

like applications). Fig. 2 shows a generic sensor set

and data sources suitable for use in ISA:

• A GNSS data source used for positioning. The

GNSS source can also be used as one of several

inputs of a multi-source localization function that

Figure 2: MBD&R example for ISA: vehicle sensors and

data sources.

enables lane-level accuracy.

• On-board cameras, as part of the perception sys-

tem, are used for Trafﬁc Sign Recognition (TSR)

of (static) trafﬁc signs and possibly overhead vari-

able message signs (VMS). Also other camera

image-based functions can be of interest, for ex-

ample line-detection for lane-level accurate ISA.

• (Dynamic) cloud data is used as a collection

term for all kinds of real-time trafﬁc informa-

tion (RTTI) services, from a multitude of service

providers, here speciﬁcally those providing speed

limits.

• (HD) Map services, next to static (ofﬂine) map

information, can include real-time map updates,

offering new layers of information suited for

ADAS/ADS functionality, such as speed limits for

ISA.

• V2X communication and connected services re-

lated to RTTI-type of data. For ISA the In-

Vehicle Information Message (IVIM), (ISO/TC-

204, 2020) is of special interest. The IVIM of-

fers speed limit and road layout information and

is normally provided by an authorized Road Op-

erator (RO).

Within our generic approach, every relevant ISA sen-

sor can be extended with MBD functionality. In

Fig. 2 the green detector blocks are speciﬁcally de-

signed for their sensor, i.e. for handling a speciﬁc

data type. Detector functions perform basic checks

on system health and status; data quality and trust

validation (consistency, integrity, plausibility, timing,

etc.); and expected misbehaviours. These individual

MB-detectors are connected to a central in-vehicle

MBD&R block that compares sensor data and cal-

culates related trust scores, calculates data quality

scores, collects evidence of misbehaviour and uses

this information for ﬁling MBD-reports. Based on

this central assessment, the speed limit data is used or

dismissed by the ISA application and detected misbe-

haviour is reported back to the providing sources. The

Quality and Trust Indicators of Digital Road Infrastructure Data Are Essential to Improve Its Usability: An Intelligent Speed Assist (ISA)

Study

633

latter facilitates the improvement of the data sources

and enhances overall system reliability.

4 VEHICLE IMPLEMENTATIONS

This section provides more detailed information on

the ISA use case and related vehicle implementations

within the DITM project. The developed concept de-

sign is being implemented into real vehicles (TNO

carlabs) and DRI-elements are deployed and used for

ISA testing under real-life conditions.

4.1 Misbehaviour Detectors

The in-vehicle MBD&R functions are depicted in

Fig. 3 and are divided into two main blocks:

• On-board Unit (OBU) stack: this is part of the

V2X communication system and does checks on

the received data. The low-level communication

function does security checks (V2X-PKI, creden-

tials, integrity), correct forwarding, data consis-

tency check, timing check etc. It can already ﬁlter

out untrusted, corrupted messages. Trusted data is

forward into speciﬁc MBD and AD functions.

• Automated Driving (AD) stack: this part contains

all in-vehicle AD related functions like sensors,

world model, drive planning and vehicle control.

For ISA it has camera and map MB-detectors and

uses the output coming from the OBU message

detectors. A central function (re-)calculates trust

scores and collects evidence to be used in MBD-

reports. Quality scores are calculated and are used

to determine which speed limit data is used as in-

put for the ISA application.

All received V2X data from DRI elements are pro-

cessed via the ”communication trust” (blue) block of

the OBU stack. This part detects generic communi-

cation misbehaviour with checks possible at all com-

munication layers (ETSI, 2010). The security layer

is the ﬁrst layer of this detector in which messages

are being assessed based on security-related checks:

trusted source, security proﬁles, integrity checks, etc.

If the message is not ﬁltered out at this stage, other

checks are performed related to message formats and

structure, consistency and timing checks. If messages

are discarded at this stage, the collected evidence can

be used to ﬁle a MB report. For trusted messages,

a trust score is calculated, upon which they are for-

warded to the ITS message detector (red block). The

ITS message detector performs checks based on the

type of message received. For ISA the IVIM is as-

sessed and speciﬁc checks on message type, expected

structure, data elements content checks are executed.

Misbehaviour checks are being performed, with spe-

ciﬁc interest in road layout information and speed

limit information at road and lane level, to verify if

data are within expected bounds, if data is consistent

over time, etc. The ITS message detector ﬁlters out

irrelevant information (for the ISA application) and

forwards the speed limit, road layout information to-

gether with a calculated trust score. In the AD stack,

all in-vehicle related Misbehaviour Detectors are per-

forming similar speciﬁc data checks and calculate in-

dividual trust scores. For ISA this are the Camera

Detector and Map Detector (red blocks). First checks

are basic system-related checks to validate healthy

operation, followed by more speciﬁc checks on sen-

sor and system data consistency, plausibility, expected

timing of signals, etc. A central block (orange) is pro-

cessing the output from all detectors to calculate an

overall trust score and to collect the evidence for de-

tected misbehaviours, which are being forwarded as

input for potential MB-reporting. In addition, qual-

ity scores are being calculated based on comparing

the different sources (Camera, Map, IVIM) which are

used to determine which speed limit to forward to the

ISA application. The quality score is, again, built-

up of several data-checks (e.g. thresholds), but also

based on comparison of sources, and use of conﬁ-

dence values (if available). In Fig. 3 the green Road

Model block uses these inputs to determine the ap-

plicable speed limit, since ultimately the application

needs a single source-of-truth speed limit to act upon.

4.2 Trust and Quality Score

Calculations

4.2.1 Trust Score

This is a continuation of our work done in (Oliveira,

2024), which already uses trust and quality scores as

part of ISA MBD functionality. A reputation sys-

tem concept is developed targeting all ISA entities, or

nodes (data sources), such as communication chan-

nels, perception systems, possibly other vehicles (via

V2X communication) etc. The used trust scoring con-

cept is based on the theory of reputation scores from

publications of (Michiardi and Molva, 2002), (Lein-

muller et al., 2008) and (Bißmeyer et al., 2012). The

trust scores are intended to track the occurrence of

misbehaviour in each node. A misbehaviour can di-

rectly be linked to a check of a detector, with a detec-

tor covering multiple checks applied in logical rela-

tions. Additionally, the interest is not only in trust at

a speciﬁc moment, but more in a historical trust over

a certain window of time. This ensures that false pos-

VEHITS 2025 - 11th International Conference on Vehicle Technology and Intelligent Transport Systems

634

Figure 3: Flow diagram of the in-vehicle MBD&R system.

itives do not suddenly raise the trust value of a node.

For a node to become trustworthy, it needs to perform

correctly for a certain amount of time. Current exten-

sions also include concepts of the Transferable Belief

Model (TBM) (Philippe Smets, 1994). Where TBM

provides a mathematical framework for rational and

evidence-based decision making under various types

of uncertainty. Like for instance inaccurate or dis-

trusted observations, indecisive or missing informa-

tion, or dealing with conﬂicts in the evidence. Appli-

cation of TBM has several features particularly rele-

vant to the detection and handling of anomalies and

misbehaviour. Evidence can be provided at differ-

ent levels, e.g. the evidence that a sensor is working,

the evidence that the sensor detected a line, and evi-

dence that the sensor recognized a trafﬁc sign, and a

sub-sign, etc. Causality of evidence can be organized

and aggregated in a complex belief model as a logical

network. Our approach is compatible with ongoing

developments covered in (5GAA, 2022) and (5GAA,

2024), which uses concepts like trust relationships,

trust referral and trust network analysis with subjec-

tive logic (J. Audun and Pope, 2006).

4.2.2 Quality Score

The quality scoring is used to evaluate the validity of

the speed limit information for use by the ISA appli-

cation. So, the quality score calculation for a single

source or node is not based on a historical evaluation

of the data. Instead, quality scoring is based on the be-

lief scores of individual sample values available from

individual Misbehaviour Detectors. To recall, for

our ISA example the individual speed limit sources

are TSR data from camera, received IVIM from DRI

and map-based speed limits. The three available iso-

lated belief scores are evaluated and undergo a ”2-

out-of-3” real-time comparison of the available speed

limit values. Depending on the belief scores, de-

ﬁned thresholds, and applied situational weighting,

the most likely speed limit is determined. Situational

weighting implies the option to take into account con-

textual knowledge by the system in computing the be-

lief (e.g. knowledge about the road type the vehicle

is currently on, with the assertion that a speed limit

sign of 30 km/h on a highway is implausible). The

quality score must be recalculated every time one of

the independent sources provides a new speed limit

value. The trust scores are calculated in real-time and

are updated per independent source. For the trust and

quality scores, computational loads and times are ex-

pected to be within acceptable limits, as the detected

speed limit changes over a certain traveled distance

are also limited. In addition, depending on individual

source trust scores or detected misbehaviours, certain

data will be discarded, or an individual source is (tem-

porarily) marked as untrusted.

4.3 Test in Real-Life Conditions

For the ISA use case the MBD&R implementations

are being deployed in our experimental carlab and test

scenarios are executed at the highways surrounding

the city of Eindhoven in The Netherlands. Within the

DITM digital data sharing environment, tooling has

been deployed to share actual IVIMs that match the

local conditions of the roads. It also supports spe-

ciﬁc predeﬁned misbehaviour into the IVIMs which

allows for replicable testing. Testing scenarios cover

the local conditions with detected trafﬁc signs, diverse

road and lane layouts, active VMS systems and com-

Quality and Trust Indicators of Digital Road Infrastructure Data Are Essential to Improve Its Usability: An Intelligent Speed Assist (ISA)

Study

635

Figure 4: Illustration of ISA MBD&R testing under real-

life conditions.

mercial map data (as depicted in Fig. 4). All ISA

relevant data is presented to the driver via the vehi-

cle HMI, but the ISA application itself is operated in

open loop, without actuation of speed control, as pub-

lic roads are being used. In addition, all relevant data

will be logged (DRI, vehicles) for post-processing and

further analysis.

4.4 Expected Results

The ISA MBD&R functions, with trust and qual-

ity scoring methods, are currently being implemented

into our carlab. Testing for veriﬁcation at vehicle level

is planned for early 2025. Thereafter real-life val-

idation can start at the highways surrounding Eind-

hoven. Relevant data will be collected for analysis

of ISA performance including speed limit informa-

tion from: maps (static and dynamic); available traf-

ﬁc signs (static signs, VMS); and lane-speciﬁc speed

limits from IVIMs. The validation data will also be

used for further improving the MBD functionality, for

example tuning the individual misbehavior detectors

and quality and trust scoring methods. Expected is

that having our MBD&R implementation deployed

will improve the performance of ISA application be-

cause of:

• the IVIM exchange via DRI offers an additional

source of ISA information.

• certain misbehaviours can now be detected.

• data sources can be identiﬁed as trusted.

• data quality and trust scores can improve usability

of the data for ISA and thus improve ISA speed

limit selections.

5 CONCLUSIONS

Current ADAS/ADS, like an ISA application can be

improved with the support of DRI. Concerns about

quality and trust in data must be addressed to fully re-

alize the beneﬁts of smart mobility. Our MBD&R im-

plementation for ISA, with cross-verifying data from

cameras, maps and extended with DRI sources, can

be a ﬁrst step in realizing such a trusted CCAM en-

vironment. This is done with ISA sources and data

identiﬁed as trusted (or untrusted) and the detection

and reporting of misbehaviours. Furthermore, with

data quality and trust scores to improve data usability

and speed limit selections for ISA. Future work in-

volves the execution of the real-life testing, scaling up

with other sensors and including other sources com-

ing from DRI. Doing more speciﬁc MBD scenario

testing, for instance by incorporating lane-speciﬁc

rules. The test results and more technical details of

the concepts will be published later in a research pa-

per. Our approach is generic and can be used for other

ADAS/ADS applications and can even be applied in

similar ways at other entities in the CCAM data chain,

e.g. at data service providers. The continuation of this

work is preferably done with participation of stake-

holders such as road operators, data providers and ve-

hicle OEMs, seeking agreements on data quality and

trust speciﬁcations to ensure effective use of shared

data.

ACKNOWLEDGEMENTS

Most of this work is done as part of the DITM project

(DITM, 2022), a National Growth Fund project subsi-

dized by the Dutch government. DITM aims to make

future mobility safer and more efﬁcient. The project

is expected to end in 2026.

REFERENCES

5GAA (2022). Misbehaviour Detection.

https://5gaa.org/content/uploads/2022/07/

5GAA-Misbehaviour-detection-Final.pdf. Page

visit on 18th of February 2025.

5GAA (2024). Creating Trust in Connected and Automated

Vehicles. https://5gaa.org/content/uploads/2024/05/

5gaa-trust4auto-white-paper-2024.pdf. Page visit on

18th of February 2025.

Bißmeyer, R., Mauthofer, S., Bayarou, K. M., and Kargl,

F. (2012). Assessment of node trustworthiness in

VANETs using data plausibility checks with particle

ﬁlters.

DITM (2022). Digital Infrastructure for

Future-proof Mobility. https://www.

tno.nl/en/digital/smart-trafﬁc-transport/

automated-vehicle-technology-public/

ditm-digital-infrastructure-future-proof/. 2022 -

2026.

VEHITS 2025 - 11th International Conference on Vehicle Technology and Intelligent Transport Systems

636

EC (2018). C-ITS Point of Contact (CPOC): European

Union C-ITS Security Credential Management Sys-

tem (EU CCMS). https://cpoc.jrc.ec.europa.eu/index.

html. Page visit on 18th of February 2025.

EC (2023). Intelligent Speed Assistance (ISA) set to be-

come mandatory across Europe.

ETSI (2010). ETSI EN 302 665: Intelligent Transport Sys-

tem (ITS); Communication Architecture, V1.1.1.

ETSI (2021). ITS; Security; ITS communications secu-

rity architecture and security management; Release 2.

Technical report, ETSI ITS.

ETSI (2023). ITS; Security; Misbehavior Reporting ser-

vice; Release 2. Technical report, ETSI ITS.

Farah, H., Erkens, S., Alkim, T., and van Arem, B. (2018).

Infrastructure for Automated and Connected Driving:

State of the art and future research directions. Techni-

cal report, Road Vehicle Automation 4. Lecture Notes

in Mobility.

ISO/TC-204 (2020). Intelligent Transport Systems, Coop-

erative ITS, Dictionary of in-vehicle information (IVI)

data structures. Technical report, OSI TC-204.

J. Audun, R. H. and Pope, S. (2006). Trust network analy-

sis with subjective logic. in Conference Proceedings

of the Twenty-Ninth Australasian Computer Science

Conference.

Kamel, J., Ansari, M. R., Petit, J., Kaiser, A., Jemaa, I. B.,

and Urien, P. (2020). Simulation framework for mis-

behavior detection in vehicular networks.

Leinmuller, T., Schoch, E., Kargl, F., and Maihofer, C.

(2008). Decentralized position veriﬁcation in geo-

graphic ad hoc routing.

Michiardi, P. and Molva, R. (2002). CORE: A Collaborative

Reputation Mechanism to Enforce Node Cooperation

in Mobile Ad Hoc Networks.

OECD (2023). Preparing infrastructure for automated vehi-

cles. Technical report, OECD / International Transport

Forum Research Reports.

Oliveira, P. (2024). Misbehaviour Detection System for In-

telligent Speed Assistance (ISA). In Proc. of 2024

IEEE Intelligent Vehicles Symposium (IV).

Philippe Smets, R. K. (1994). The Transferable Belief

Model. Artiﬁcial Intelligence.

van der Heijden, R. W., Dietzel, S., Leinmuller, T., and

Kargl, F. (2019). Survey on misbehavior detection in

cooperative intelligent transportation systems.

Zhang, J., Jemaa, I., and Nashashibi, F. (2022). Trust man-

agement framework for misbehavior detection in col-

lective perception services. In 17th International Con-

ference on Control, Automation, Robotics and Vision

(ICARCV).

Quality and Trust Indicators of Digital Road Infrastructure Data Are Essential to Improve Its Usability: An Intelligent Speed Assist (ISA)

Study

637